Microsoft Unveils Azure AI, Blockchain And IoT Tools Ahead Of Build 2019

‘Since the inception of Azure, we have been focused on delivering a true hybrid cloud where applications spanning public cloud and on-premises data centers are built and run consistently,’ says Scott Guthrie, executive vice president of Microsoft Cloud and AI Group. ‘As organizations are now building applications that span the intelligent cloud and intelligent edge, the same approach is needed.’

Microsoft is unveiling new Azure artificial intelligence, mixed reality, Internet of Things and blockchain services and tools for developers to create hybrid applications for cloud and edge computing.

The announcements come ahead of Microsoft’s Build 2019 developer conference that starts in Seattle on Monday.

“Since the inception of Azure, we have been focused on delivering a true hybrid cloud where applications spanning public cloud and on-premises data centers are built and run consistently,” Scott Guthrie, executive vice president of the Redmond, Wash., company’s Microsoft Cloud and AI Group, said in a blog post. “As organizations are now building applications that span the intelligent cloud and intelligent edge, the same approach is needed.”

Artificial Intelligence And Machine Learning

Artificial intelligence (AI) use is growing rapidly across Azure, according to Microsoft. More than 1.3 million developers use Azure Cognitive Services, the No. 2 cloud provider’s set of application programming interfaces, software development kits and container images that allows them to integrate ready-made AI directly into their applications. They create almost 3,000 new bots each week.

New Azure AI technology includes the expansion of Azure Cognitive Services, which currently includes text analytics, natural language understanding, facial detection, and speech and vision recognition capabilities.

“Azure is the only public cloud that enables these Cognitive Services to be containerized to run on-premises, in the cloud and at the edge,” Guthrie said.

A new Decision category includes services that not only observe the world, such as image recognition, but build in a sophisticated decision process, according to the company. Microsoft Azure is previewing the Personalizer service under this category, which uses reinforcement learning to provide users with a specific recommendation to allow for quick and informed decision-making.

In its Vision category, it’s unveiling two new services in public preview at Build. Ink Recognizer enables developers to combine the benefits of physical pen and paper with the best of the digital by embedding digital ink recognition capabilities. Form Recognizer automates data entry by extracting text, key value pairs and tables from documents.

A new Speech service, Conversation Transcription, is an advanced speech-to-text capability that transcribes meeting conversations in real time to allow participants to fully engage in discussions, know who said what and when, and more quickly follow up on next steps.

In the Language category, Language Understanding will have a new analytics dashboard to evaluate the quality of language models, and Q&A Maker will include the added ability to create multi-turn dialogues.

Microsoft Azure, which is continuing to make more Cognitive Services available on the edge and on-premises through the use of containers, is introducing new container support for speech to text, text to speech, form recognizer and anomaly detector.

Microsoft is incorporating AI into Azure Search with the general availability of a cognitive search capability, which allows customers to apply Cognitive Services algorithms to extract insights from their structured and unstructured content. It’s also previewing a new capability that allows developers to store AI insights gained from cognitive search to apply to scenarios such as visualizations for Power BI, its business analytics service, or machine learning models.

“Furthering our commitment to building the most productive AI platform, we’re delivering key new innovations in Azure Machine Learning that simplify the process of building, training and deployment of machine learning (ML) models at scale,” Guthrie said.

Updates include a new visual ML interface that provides a zero-code model creation-and-deployment experience using drag-and-drop capabilities, and an automated ML capability that makes developing high-quality models easier.

New MLOps capabilities with Azure DevOps integration will provide developers with reproducibility, audit ability and automation of the end-to-end machine learning life cycle, according to Microsoft spokesman Frank Shaw.

“With ML Ops, we are bringing the best of the DevOps world to machine learning to enable debugging and deployment, model validation, profiling, monitoring and (continuous integration and continuous delivery,) leading to efficiency and improved model quality over time,” he said. “These capabilities will help organizations of all sizes quickly and efficiently put the work of data science teams into production at scale.”

Microsoft Azure also announced the general availability of hardware-accelerated Azure Machine Learning models that run on field-programmable gate arrays, and Onyx run-time support for Nvidia, TensorRT and Intel nGraph for high-speed inferencing on Nvidia and Intel chip sets.

Mixed Reality

Microsoft is expanding its Mixed-Reality Developer Program.

“Over the past year, we have engaged with more than 20,000 mixed-reality developers, a number we expect to more than triple in the coming 12 months,” Julia White, who leads product marketing at Microsoft Azure, wrote in a blog post. “To serve the needs of a larger audience, we are investing in more meetups, programs, events and hacks -- including our Mixed Reality Dev Days that is happening now -- with an increased focus on serving the mixed-reality dev community on a global scale.”

To help developers create applications for HoloLens 2, its mixed-reality smart glasses technology, Microsoft announced the HoloLens 2 Developer bundle, which gives them access to solutions to build and run mixed-reality experiences across a range of devices.

“We have barely scratched the surface for possibilities with mixed reality development,” Guthrie said.

Unreal Engine 4 support for HoloLens 2 will become available by the end of this month to support developers with streaming and native platform integration to create high-quality, photo-realistic renders and immersive mixed-reality experiences for solutions including architecture, product design and manufacturing.

Azure SQL Database Edge

Microsoft Azure is extending its data storage and analysis capabilities to the edge with its private preview of Azure SQL Database Edge, a SQL engine that runs on ARM processors and is optimized for lower compute requirements, with built-in AI that combines data streaming and time series data with in-database machine learning and graph capabilities.

“And because Azure SQL Database Edge shares the same programming surface with SQL Server, you can easily take your applications to the edge without having to learn new tools and languages, allowing you to preserve consistency in application management and security control,” Guthrie said. “This consistency in database programming and control plane across cloud and edge is essential to running a secure and well-managed hybrid application.”

Internet of things and blockchain

To help customers challenged by deploying internet-of-things (IoT) solutions at scale, Microsoft is introducing IoT Plug and Play, a new open modeling language to seamlessly connect IoT devices to the cloud.

“Previously, software had to be written specifically for the connected device it supported, limiting the scale of IoT deployments,” Guthrie said. “IoT Plug and Play provides developers with a faster way to build IoT devices and will provide customers with a large ecosystem of partner-certified devices that can work with any IoT solution.”

Microsoft said it’s doubling down on blockchain investments after last year announcing Azure Blockchain Workbench, which gave developers a simple user interface to model blockchain applications on a pre-configured, Azure-supported network.

Next up is Azure Blockchain Services, which is designed to simplify the formation, management and governance of consortium blockchain networks so businesses can focus on workflow logic and app development. It deploys a fully managed consortium network with a few clicks and offers built-in governance for common management tasks such as adding new members, setting permissions and authenticating user applications.

The service already is getting enterprise traction, according to Guthrie.

“We also announced this week that J.P. Morgan’s Ethereum platform, Quorum, is the first ledger available in Azure Blockchain Service, enabling Microsoft and J.P. Morgan to offer the first enterprise-grade Ethereum stack to the industry,” he said.

Microsoft’s new blockchain extension for VS Code lets developers create and compile Ethereum smart contracts, deploy them to either the public chain or a consortium network in Azure Blockchain Service and manage their code using Azure DevOps.

"The ledger is just the foundation for new applications,” Mark Russinovich, Microsoft Azure’s chief technology officer, said in a blog post. “After configuring the underlying blockchain network with Azure Blockchain Service, you need to codify your business logic using smart contracts. Until now, this has been cumbersome, requiring multiple command-line tools and limited developer IDE integration.”

Microsoft also is releasing a new Azure Blockchain Dev Kit with connectors and templates for Logic Apps and Flow, and integrations with serverless tools such as Azure Functions.